Recording line level audio

[Mike Burke]

I've just received our camera back from Red with the audio upgrade.
Using release Build 17, upgraded audio board, with new cables.
Audio menu settings are as follows: Line level, 0db.

What I am finding regarding line level audio levels is totally confusing.
With a mixer supplying zero level tone, at line level,
the LED display shows levels well into the orange quadrant of the display.
This is to the right of the solid line.
Having lowered the output of the mixer to the solid line of the
camera LED display, my mixer now displays an output of -10db.
Clearly something is out of cal.

Exactly what levels are these hash marks on the LED display representing?
There is no information in the manual.
What is the proper procedure for establishing a -20db input level
when using a 0db reference tone from a mixer.
Any one from Red?

This camera begins a commercial job Friday, utilizing dual audio recordings.
Thanks much.

[Stuart English]

Process is just this-

Feed the camera - 20dB FS with Line Input gain set to 0dB. The -20dB FS point should line up to the right hand hash mark.

Make a recording and verify in FCP or other software application that you are at -20 dB.

[Mike Burke]

Process is just this-

Feed the camera - 20dB FS with Line Input gain set to 0dB. The -20dB FS point should line up to the right hand hash mark.

Make a recording and verify in FCP or other software application that you are at -20 dB.

Stuart, thanks for the response.

My quandry is this: In order to achieve -20db FS at the camera,
I must set my mixer output to somewhere around -10 VU, with tone as a reference.
Something must be wrong that I'm overlooking.

[Joel Kaye]

Is this any different on build 18?

[Stuart English]

My quandry is this: In order to achieve -20db FS at the camera,
I must set my mixer output to somewhere around -10 VU, with tone as a reference. Something must be wrong that I'm overlooking.

Two questions -

How are you measuring that?

Are you using the mini-XLR to XLR cable or the cable with mini-XLR to mini-XLR plus mini-XLR to XLR adaptor?

As to the Build 18 question: its the same as 17.

[Mike Burke]

Is this any different on build 18?

I can't say as I haven't tested Beta 18 yet. I probably won't get
the opportunity to test 18 by the time this job starts.
What I can say is this has changed as a result of the board upgrade, or
the change to Build 17. I had been on Build 16 up until that point.

I'm really curious as to what others with the new board have experienced.

[Mike Burke]

Two questions -

How are you measuring that? Are you using the mini-XLR to XLR cable or the cable with mini-XLR to mini-XLR plus mini-XLR to XLR adaptor?


1. I'm referring to the VU display on the mixer. Actually, I used a second mixer
to confirm that the mixer wasn't the problem.

2. I've used both: mini-XLR to XLR cable as well as the cable with mini-XLR to mini-XLR plus mini-XLR to XLR adaptor.

[Stuart English]

O.K. The VU on the mixer will tell you the output level of the mixer itself, so lets assume that is fine at 0dB / - 20 dBFS. The reason I ask about the type of XLR input cable you are using is that the mini-XLR to XLR version has a -10dB pad in it to help the older style audio board accept higher input voltage levels than it otherwise could. In the new audio board, those pads are designed into the audio pre-amp so those cables shouldn't be used, else you will over attenuate.

If you were using -20dBFS out of your mixer, and mini-XLR to mini-XLR cable with mini-XLR to XLR adaptor, I'd expect the level being recorded to be fine.

[Mike Burke]

O.K. The VU on the mixer will tell you the output level of the mixer itself, so lets assume that is fine at 0dB / - 20 dBFS. The reason I ask about the type of XLR input cable you are using is that the mini-XLR to XLR version has a -10dB pad in it to help the older style audio board accept higher input voltage levels than it otherwise could. In the new audio board, those pads are designed into the audio pre-amp so those cables shouldn't be used, else you will over attenuate.

If you were using -20dBFS out of your mixer, and mini-XLR to mini-XLR cable with mini-XLR to XLR adaptor, I'd expect the level being recorded to be fine.

I should have specified better.
The mini XLR to XLR cables are custom made at our studio.
They have no pads whatsoever. The input level at the camera is identical
to that of the mini-XLR to mini-XLR cable with mini-XLR to XLR adaptor provided by Red.

It's odd. At this point, I wish I still had one of the padded cables to compare with.
My level seems to be about 10db too hot.

[Stuart English]

re "10 dB too hot .. "

You mean the level you need to send it / the level that gets recorded / or the level the camera sends back?

[Mike Burke]

re "10 dB too hot .. "

You mean the level you need to send it / the level that gets recorded / or the level the camera sends back?

With a mixer supplying tone at zero VU, line level,
the LED display shows levels well into the orange quadrant of the display.
This is to the right of the solid line.
In order to achieve a level at the camera of -20db FS, (at the solid line),
I must attenuate the output of the mixer -10db.
The camera audio inputs are set at line level, 0db.
I am using cables that have no pad, supplied by Red.

[Mike Burke]

With a mixer supplying tone at zero VU, line level,
the camera LED display shows levels well into the orange quadrant of the display.
This is to the right of the solid line.
In order to achieve an input level at the camera of -20db FS, (at the solid line),
I must attenuate the output of the mixer -10db.
The camera audio inputs are set at line level, 0db.
I am using cables that have no pad, supplied by Red.

Has anyone seen anything like this, or have all line level audio
records with the audio board upgrade been successful?
We're pretty baffled here...

[Tom Visser]

what mixer are you using?

[Mike Burke]

what mixer are you using?

Thanks for the reply.
I've confirmed this condition with (2) Shure FP-33, as well as a Mackie 1402.
All yielded the same results.

[Stuart English]

With a mixer supplying tone at zero VU, line level,
the LED display shows levels well into the orange quadrant of the display.
This is to the right of the solid line.
In order to achieve a level at the camera of -20db FS, (at the solid line),
I must attenuate the output of the mixer -10db.
The camera audio inputs are set at line level, 0db.
I am using cables that have no pad, supplied by Red.

Yes I understand that. My question then is if you ignore the PPM meter of the camera and record the -20dB FS signal to the camera with 0dB Gain applied, what is the level of the recorded signal? i.e is - 20dB FS also?

The way I check that is to bring the recording into Final Cut Pro and observe the level in the audio adjust panel.

[Tom Visser]

I think your problem lies in your audio input level settings on the RED. By setting to Line at 0dB, you are saying that full scale or 0dBFs would be 0dBU. Your mixer has a headroom of somewhere in the range of +15dBU to +18dBU. Try setting the line input to +18dBU and then see if that allows your meters to match up correctly. If not, just adjust your line settings until it matches. Make sure you remember to set the outputs on the FP33 back to full range.

[Stuart English]

I think your problem lies in your audio input level settings on the RED. By setting to Line at 0dB, you are saying that full scale or 0dBFs would be 0dBU..

With respect, that is not true, so don't attempt to do that. Please do the test I described first ?

Thanks,

[Mike Burke]

Yes I understand that. My question then is if you ignore the PPM meter of the camera and record the -20dB FS signal to the camera with 0dB Gain applied, what is the level of the recorded signal? i.e is - 20dB FS also?

The way I check that is to bring the recording into Final Cut Pro and observe the level in the audio adjust panel.


Alright, I've done as you've described regarding the input level to the camera,
disregarding the PPM meter. I've made a recording applying -20db with the camera input set to line level, 0db.

While I don't have access to a Final Cut system at the moment, I was able to perform another test.
I've played out of the Red Cam at 1080P, with audio embedded on the SDI to an HD-D5.
With the VTR audio input in the unity or fixed mode, calibrated for line level, -20dbFS,
the recorded level on the camera hits the VTR at -8db.
This appears to 12db hotter than it should be.
While I will try the FCP test,
this seems to support everthing I have found up until now.
And also, the recording is mildly overmodulated.

[Tom Visser]

I think your problem lies in your audio input level settings on the RED. By setting to Line at 0dB, you are saying that full scale or 0dBFs would be 0dBU. Your mixer has a headroom of somewhere in the range of +15dBU to +18dBU. Try setting the line input to +18dBU and then see if that allows your meters to match up correctly. If not, just adjust your line settings until it matches. Make sure you remember to set the outputs on the FP33 back to full range.

ah, looking at the manual on page 10 build 16, it led me to believe that the line input was adjustable, but then on page 11 it mentions that it is not. my bad.

[Stuart English]

Alright, I've done as you've described regarding the input level to the camera,
disregarding the PPM meter. I've made a recording applying -20db with the camera input set to line level, 0db.

While I don't have access to a Final Cut system at the moment, I was able to perform another test.
I've played out of the Red Cam at 1080P, with audio embedded on the SDI to an HD-D5.
With the VTR audio input in the unity or fixed mode, calibrated for line level, -20dbFS,
the recorded level on the camera hits the VTR at -8db.
This appears to 12db hotter than it should be.
While I will try the FCP test,
this seems to support everthing I have found up until now.
And also, the recording is mildly overmodulated.


Build 17 added an option for + 10dB gain on line inputs, can you verify that that is not enabled.....

[Mike Burke]

Build 17 added an option for + 10dB gain on line inputs, can you verify that that is not enabled.....

Yep. Done that.

[Brian Ferguson]

When I take a 702T Sound Devices, go into the recorder's menus and select -20 dBFS on 1Khz tone and output from the TA3 line out to the TA3 line in on the RED One the resulting playback in FCP is -20 dBFS.

On my 302 Sound Devices mixer the 1K tone at 0dBU is just slightly hotter like -18 or -19 dBFS when played back but really close to -20. You have to make sure your camera line level is set at 0dBFS and not +10 under the Video/Audio/Line.

[Kip Hewitt]

I too am having this problem with a newly upgraded audio board.

On a shoot this weekend we blasted the line level with a Mackie Mixer until it metered just a bit under peaking.

Playback in camera seemed to sound fine.

But when played back in Final Cut Pro or Quicktime, the audio is whisper quiet. The FCP audio meters barely get above the -30 marker.

Yes, I am using the correct cable and yes my line level is set to 0db.

Thank you.

[Steve Sherrick]

I ran into this recently as well when using a feed from a Shure Mixer on set. I'm about to run a little test and will confirm for everyone what the deal is. Be back in a little bit.

[Jeremy Neish]

We are also experiencing this problem on our upgraded camera. We've verified all of the above and are still over modulating.

[HD Hildebrand]

I ran into this problem as well.

[JonathanF]

We saw this on a recent shoot as well and ended up going right to the external decks and bypassing the camera altogether. Not my first choice as now I have a bunch of added cost to make sure all our audio is lined up when we post when I should have at least had reference on camera. When the audio got into FCP the levels were to quiet to be usable.

We also had the studio's Sound Devices mixer and thank god we decided only to use the on camera sound only as reference (we went off the outputs on the deck to the mixer, then out wireless to the camera).

While I'm not a sound guy and I don't pretend to play one on TV, it would be REALLY nice of someone would actually do some work on this area of the docs, our sound guy (a veteran who just finished work on an Academy Award winning film) can feel a little better about working with a camera system we're otherwise pretty comfortable with. His words were precisely, "these docs are totally worthless."

This shouldn't be such a mystery, but as with most things, it boils down to poor technical documentation (I have to agree with our soundman), and relying on a web forum to fill the gaps. That's a dangerous choice in my opinion as it's not accessible in the field all the time, and is a poor (or worse, misleading) reference in a pinch when you just need to know one little technical detail in a hurry (while a producer is breathing down your neck on set wondering when he can get the take).

While it's true these things are best resolved in pre-production, there are always little questions that arise on set (why my ASC manual is always in my kit when I am filming, I don't have hyperfocal charts for every format dedicated to memory and I can't expect the soundman to have the technical details of every camera system in his memory banks either).

To be blunt, there are too many morons on the internet trying to prove their intelligence to the universe to rely on web forums as a form of technical documentation. I think this is one of the little areas Red can focus on to improve the customer experience with their products (which we otherwise love).

The forums are a nice thing for immediate support and for building community, but they will never replace proper technical manuals or documentation (this includes the software products as well which really suffer in this area).

I think while the camera probably does the right thing, it would be nice if some effort was put into making some extensive updates to the documentation and getting that out into everyone's hands asap.

J

[Zakaree Sandberg]

upgraded audio board.. all kinds of issues..

tone peaking...
evf flashing when recording audio
(think i figured the last one out.. happens with NON RED xlr-xlr mini cables...
or when recording 2 channels.. dont know which one yet)

[Allan Stallard]

This is what I got from my trusted audio guy. I hope that this helps clarify and I hope it's all accurate. Please comment on inaccuracies as we are going into a 4 day shoot Friday.

"The 0db level on the red is neither -10dBv (consumer line) nor is it +4dBu (professional line) it is 0dBu (another line level that appears on some mixers like a Mackie, and also on DVCPro decks). This level is in-between the other two line levels. So… sending a tone at 0dBu out of, say a Mackie which will show tone at 0dBu on the meter, should come up to -20dBfs OR whatever the RED is set up to receive that 0dBu signal. Other manufacturers use -16dBfs, -18dBfs at 0VU reference level on a +4dBu or 0dBu line level.

It is important to remember that the dBfs scale does not translate directly to digital meters showing VU, and those VU meters can be set to any of the three line level signals (+4dBu, 0dBu, -10dBv). Also, there is not 14dB difference between +4dBu and -10dBv, nor is there 10dB difference between 0dBu and -10dBv. They are different scales based on different voltages. It is advisable to use a 0dB scale like on a Mackie to check your tone and where it comes up on the RED."

[JonathanF]

Considering we have a 2 camera shoot for a never to be repeated event in just three weeks, using two RedOne's with the new revision audio board, it would be of some interest if someone from Red would actually make a comment on this.

Our sound guy called me again today asking if there had been any official word on this.

Seems making some clarifying additions to the manual would be a really good step in the right direction for a lot of us who would like to take better advantage of the improved sound capabilities.

Not trying to be an a$$, but we can really use some official clarification here, and not just in the forum, I'm taking about actual documentation I can put in the hands experienced crew (who have what I think is a justifiable beef with the extent of documentation on the sound subsystem for the camera).

Jim... anyone?

Cheers,

J

[Jeremy Neish]

This thread has gone awfully quiet but seems to be lacking resolution. Has anybody received any answers from the powers-that-be? Or figured out a solution on their own? I submitted a tech support request to RED but never received a reply. We really need an solution to this issue.

In summary, we continue to see line levels that are much hotter than they should be, (levels on our Mackie mixer are barely registering but are very hot on the camera) and even if we turn the mixer way down to compensate, the audio imports distorted with clipped waveforms, even though the meters in the camera never clipped. I've triple checked and the camera is set to line level (0db).

[JonathanF]

Jeremy,

I posted this:

http://www.reduser.net/forum/showthread.php?t=27282

And got a bit of love, but we need to docs updated still, so we're holding our breath for the moment.

J

[Jeremy Neish]

Interesting thread, thanks for the link. But at least in my experience, the issue seems to be more than just insufficient documentation. No matter what we do, (including trying several different mixers), line level audio is distorting in our camera, even at very low levels.

[JonathanF]

Agreed, there may be other issues at play here. But until we have technical specifications and documentation we're not going to know what we need to do to test accurately.

Once you have the docs you should be able to create two or three valid test cases and know, with near certainty what your results will be.

If they differ from your expected result then you can say to the engineers, I did this and I expected this result, but got this instead; why?

Otherwise you're just tossing darts right?

I just want to be able to generate a suite of test cases record the results after confirming correct operation.

This is important for any production system you may be using in the field. If something starts behaving badly or outside of the expected operational parameters you have a handful of (usually quick) tests you can perform to confirm your suspicion. You'll know if you do x, you should get y result. If not, you got a problem, and hopefully you know where.

J

[HD Hildebrand]

Interesting thread, thanks for the link. But at least in my experience, the issue seems to be more than just insufficient documentation. No matter what we do, (including trying several different mixers), line level audio is distorting in our camera, even at very low levels.

I agree. To be honest, if I had the opportunity I'd take my old sound board back. There may be noise on phantom, but I had a custom cable made where I took the phantom out of one channel, but the actual audio out of a different channel. Result clean sound. Now, as discussed, distortion/seeming over-modulation even though line levels were in range and levels when imported into FCP are -12 or less.

I'm sorry, but this needs to be fixed.

[Brian Ferguson]

I think part of the the problem is that an analog mixer at 0 dBu is not the same as sending a -20 signal from a digital device like a Sound Devices recorder which you can set the menu at a specific tone level.

If I send a signal from a digital source at -20 dBFS, out at analog +4 line level from a 702T it hits the right hash mark. It is a simple TA3 to TA3 connection.

If I send a 0dBu 1k reference from my 302 Mixer it is slightly hotter. I am sure every analog mixer is going to send a different level. Early on back in build 12 when audio on Red was new, I was on a shoot where a PSC mixer was sending a level that had to be 30db hotter than my 302. And that was supposedly standard line level. So I know mixer to mixer the 0 dBu tone will vary, it isn't the camera.

If you send a known digital to analog line level signal like the first example (the 702) it will record and open up in FCP where it should. The analog source can be within a range. Analog mixers are all over the place. My 302 comes really close but it is not as exact as using the 702T.

And as to the Mackie references in this thread do they send a 0dBu 1k tone? I know none of my Mackies send tone, but they are older.

[Ivan Cortazar]

If I send a 0dBu 1k reference from my 302 Mixer it is slightly hotter. I am sure every analog mixer is going to send a different level. Early on back in build 12 when audio on Red was new, I was on a shoot where a PSC mixer was sending a level that had to be 30db hotter than my 302. And that was supposedly standard line level. So I know mixer to mixer the 0 dBu tone will vary, it isn't the camera.

Thanks for your explanation. Forgive my ignorance since I am not a sound person. I have a Sound Devices Mixpre which I think it should be very similar to the 302. So do you think it would be able to get good sound after the soundboard upgrade? So how do you do it? Because you can't adjust in the camera. When in Line mode, you can only put 0db or +10 Db. I guess since you are saying that the 302 comes a little bit hot, you will have to adjust in the master volume in the Mixpre and have it always out a little lower than what it should. But that is not very precise and I bet a sound guy would be quite annoyed. How much lower should he go before it peaks?.

Thanks

[Jason Hemmerlin]

Hello all,

I'm a sound guy who is new to the Red. I have a feature coming up in a week that I will be required to mix to the camera for for the on set audio. I have heard a million horror stories with this cameras audio section. I have pushed for a dual system recording and for once production listened to the Audio Department as we will be renting a 744t. My original plan was to record the isolated tracks on the 744t and record the mix to camera, but readin all of this I think I am having second thoughts about this decision.

I have noticed alot of people bringing up the 302 mixer. I own one along with a 442 as well. I've made a point of learning everything I can about the equipment I use. This has led me to believe that those of you who have a 302 mixer are not taking full advantage of it's features. If you are getting levels at the camera that are too hot at line level you do have the option to attenuate the output of your 302 from within the menu of the mixer. If it seems hot than lower the output of the mixer in the mixer menu by the amount that seems best.

Another Sound guy I recently worked with mentioned working with the Red and said that the best audio he has gotten out of that camera was by setting the nominal input level of the Red to mic level -20 db and using a -10 db level on the output of his mixer ( which is easily reached by both the 302 using the mixer menu, and the 442 by simply using the switches by the ouptut section of the mixer). Anyone using a mix-pre will need to use pads at the output to do this. Doing this would simply mean that you're driving the mic level inputs hotter than normal. He did this because going line level was too quiet. Which goes along with the too hot of a level on the meters. It seems to me that these new boards are not properly calibrated with the meters on the camera and thus the levels are off based on the way people set them up.

I will now be requesting a test with the camera that we are shooting with for this feature, as I do not want to jeopardize my career for someone else's vision.

If I find anything questionable or anything that works I'll post.

Thank You,

Jason Hemmerlin

PS- If this camera is supposed to revolutionize the non-linear video world and possibly replace the film camera, then why the hell did they bother putting an audio section on it. Leave the audio to the Sound Dept. and the video to the Camera Dept.

[Jason Hemmerlin]

Red shold also take a hint from all of the other manufacturers out there and make the inputs switchable giving a choice of only mic or line level and then having a gain control that you can fine tune the levels with. It's been a standard way of setting up audio on cameras for years. If it ain't broke don't fix it.

[Stuart English]

Jason, that's exactly the way it works.

You can choose Line or Mic level, and each Mic level input has an adjustable pre-amp gain.

The only difference is on Line we don't give you adjustment. You set the levels on your mixer.

[Jeremy Neish]

PS- If this camera is supposed to revolutionize the non-linear video world and possibly replace the film camera, then why the hell did they bother putting an audio section on it. Leave the audio to the Sound Dept. and the video to the Camera Dept.

Because this camera is designed to be flexible enough for more than just film style shooting. There are many times when I am using the camera without a sound department at all (in fact most of my shoots which are industrial, ENG or one man shoots, and have a dedicated sound department). Sometimes I get a little tired of so many folks viewing the camera as just a film-replacement camera, I see it as much more.

[Scott C.]

... then why the hell did they bother putting an audio section on it. Leave the audio to the Sound Dept. and the video to the Camera Dept.

I can think of a couple reasons:

1. For projects like Jeremy's a direct feed to camera is practical, efficient and adequate. This group is a large part of the camera's target audience.

2. Dailies- having synched scratch audio embedded into the files makes turning out dailies quicker. I'm guessing that assistant editors love this.

You seem worried but don't fret; this doesn't threaten sound jobs. Good sound techs are still required to capture a good feed whether it is going only to camera, or also to outboard multitrack.

[Jon Frost]

I would strongly advise going 2 system audio with the RED build 17... You can still get decent audio at Line Level by adjusting the output from a SD442 Mixer and the input of the RED. You have to experiment (before actual production of course) with your mics/mixer/camera settings to obtain the best audio. I still think it makes sense to have 24-bit 48K audio recorded on an SD or Deva recorder sunc with a smart slate and planning for the time to sync dailies until RED can come up with a stable audio card/recording system that can handle the same quality of a Sound Devices or Deva recorder without the hassles.

[Meryem Ersoz]

PS- If this camera is supposed to revolutionize the non-linear video world and possibly replace the film camera, then why the hell did they bother putting an audio section on it. Leave the audio to the Sound Dept. and the video to the Camera Dept.

Uh, because they listen to their customer base. You're a tad late to the party....

I just sent in one of my cameras for the audio board upgrade, and this thread has me a tad concerned...and confused. Oh well, I guess I'll refer back to it once I have a chance to test the new board.

[zach nasits]

I am trying to figure out whether or not to upgrade the audio board on my Red One. We were contacted by Red to let us know that we are up for the "RED One upgrade program." A lot of the program sounds great! However, I am hesitant to upgrade the "audio boards."

My reasons are as listed:

1) Our sound mixer has had problems working with the new audio boards and says that we should "keep ours as is!" We currently have great audio recording to the camera, via mixer and some -20db pad cables. Our sound mixer says that with the new audio boards, he insists on recording dual-source and has now incorporated -30db padded cables.

2) 2 of our other sound mixers say that they still prefer the original audio boards, now that they know how to "work them."

3) I am a cameraman and listen to my sound mixers who have provided me great sound so far!

I want to hear about other experiences. Anyone?

And to the RED Bomb Squad (with much respect), may I send in my RED ONE under the following circumstances?

a) for the i-pin install and "a tuneup that will include full camera diagnostics, re-seating of joystick knob, upgrading CF reader/writer to current gen with 16 GB CF card support, replacing many hex screws with Torx Plus screws, cleaning of OLPF and ventilation chamber, etc?"
b) leaving the original audio boards and follow my "if it ain't broke don't fix it" philosphy?

Thank you RED for producing a digital cinema package that I love to work with.

Zach Nasits
RedONE (2501) /MKII's / Angenieux HR / 2060

[Tom Visser]

I think part of the the problem is that an analog mixer at 0 dBu is not the same as sending a -20 signal from a digital device like a Sound Devices recorder which you can set the menu at a specific tone level.

If I send a signal from a digital source at -20 dBFS, out at analog +4 line level from a 702T it hits the right hash mark. It is a simple TA3 to TA3 connection.

If I send a 0dBu 1k reference from my 302 Mixer it is slightly hotter. I am sure every analog mixer is going to send a different level. Early on back in build 12 when audio on Red was new, I was on a shoot where a PSC mixer was sending a level that had to be 30db hotter than my 302. And that was supposedly standard line level. So I know mixer to mixer the 0 dBu tone will vary, it isn't the camera.

If you send a known digital to analog line level signal like the first example (the 702) it will record and open up in FCP where it should. The analog source can be within a range. Analog mixers are all over the place. My 302 comes really close but it is not as exact as using the 702T.

And as to the Mackie references in this thread do they send a 0dBu 1k tone? I know none of my Mackies send tone, but they are older.

You can't really send a 0dBFs tone out of a mixer, unless your interconnection is digital. Once it comes out of an analog line output, it is now in the dBU power range, and the reference to dBFs where 0 is digital peak, no longer has any relevance. You can set you tone generator on something like the 702 to -20dBFS and if your outputs are +4dB, you happen to match up at the camera, great. It is actually great foresight on your company's part to make the Sound Devices levels the standard. As pointed out, there is the ugly truth that not all audio devices have agreed to use the same power level outputs for reference.

The 702t / 302 difference might simply be different ballistics of the meters or performance of the tone generator itself, the 7## being a digitally produced sine wave and the 302 using an analog oscillator. The advanced settings on the 302 have quite a few output adjustment options "hidden" in a documented setup mode.


Jason, that's exactly the way it works.

You can choose Line or Mic level, and each Mic level input has an adjustable pre-amp gain.

The only difference is on Line we don't give you adjustment. You set the levels on your mixer.

Here I would recommend that the camera should really have either a stepped or continuously variable attenuation pot on the camera. The whole point of calibration is to calibrate the downstream device to the upstream device. Maybe the mixer is a little bit out of spec, so you can use the pots to make a small adjustment to the levels. Maybe the audio has to go through several hundred feet of cabling and is a little attenuated. Maybe there is a distribution amplifier that is not quite unity gain and is boosting the levels a dB or two. In any event, I would recommend that fine adjustments should be available on the camera or any recording device. Trying to establish a standard on a camera is a bit dicey, as there are US standards, EU standards, and different standards based upon the era that a mixer was manufactured and the philosophy of the designers at that point in time.

It sounds like a few people on this thread might simply have defective boards. Especially those people that are reporting hot audio levels on the meters and actually experiencing -30dB or lower levels on the workstation.

I don't like to do this, but sometimes I have to set my 302 padded down to mic level and then use the mic inputs on the camera. Conceptually this gain strategy would be at a disadvantage to sending a hot line signals, but in fact some cameras perform better with mic inputs than line inputs. I don't think that the REDs that I have worked with, both old and new boards, have every been fed anything other than line level by me, and haven't had major problems with either version of the board once I knew what to look for. I have seen "damaged" boards where different inputs seemed to be all over the place with respect to their own levels relative to each other. Headroom can also be limited (perhaps due to electronic damage), but to play it safe, I always recommend recording double with scratch audio sent to the camera. Every effort is made to keep the scratch audio of the highest quality possible, but it keeps my mind at ease being able to know that I can do the job asked of me with as few technical glitches as possible. Camera department appreciates me staying away from their camera too, I'm sure, as my scratch audio is always sent via a wireless hop.

[James Brundige]

Just received my upgrade and ran a line level test. -20 tone from Sound Devices, line input to Red at 0dB. (the other option is +10)

tone hit the hash mark, maybe a touch hot. Audio was fine as long as I did not exceed -20 out of the mixer. Not much headroom above that. Got pretty crunchy if there was any red on the Sound devices meter, which corresponds to yellow on the Red Camera meter.

The test tone reads out as -16 in FCP, which is -28 below unity.

[Stuart English]

How much headroom are you estimating James ?

No comment on FCP audio level scale... we know a -20dBFS tone input delivers a -20dBFS output..

[James Brundige]

How much headroom are you estimating James ?

No comment on FCP audio level scale... we know a -20dBFS tone input delivers a -20dBFS output..

I've got a shoot today with Red audio. I'll test some more and give a more clear answer.

About FCP, I'm confused about their meter. It tops out at +12, which I think is 0 to the rest of the world. What's dBFS?

[David Limpus]

dbFS is dB Full Scale, maximum of 0dbFS.
Reference level is -20 dBFS or -18 dBFS depending where you are in world and what standard you work to.

[James Brundige]

Another quick test while loading the van. This time a Wendt X3 mixer. 0dB tone from the mixer just lights the yellow in the camera.

In FCP, reads out as -14, -26 below unity.

Ran some dialogue tests. It seems you've got about 6 dB headroom after the camera meter hits yellow.

I can not run the Wendt, or the Sound Devices, at optimum levels. Normal output distorts the Red. If I run those mixers conservatively, it sounds fine. Since it's digital, the noise floor is low. Not ideal, but usable for now. Why not +4dB ????

Conclusion: Be very careful with the Red one meter. Keep it green. Yellow is dangerous and red is cooked.

[Stuart English]

...... a Wendt X3 mixer. 0dB tone from the mixer just lights the yellow in the camera.

Ran some dialogue tests. It seems you've got about 6 dB headroom after the camera meter hits yellow.

I can not run the Wendt, or the Sound Devices, at optimum levels. Normal output distorts the Red. If I run those mixers conservatively, it sounds fine. Since it's digital, the noise floor is low. Not ideal, but usable for now. Why not +4dB ????

Conclusion: Be very careful with the Red one meter. Keep it green. Yellow is dangerous and red is cooked.

That suggests your input level is too high or you have +10dB gain enabled for line input.

It shouldn't be lighting up yellow, as you are correct when it changes from green to yellow its indicating about 6dB of remaining headroom before clip...

When I run my Sound Devices 302 (I don't have it with me unfortunately) at -20dBFS, my camera indicates its hitting the witness mark right on the nail.....


What build of firmware are you on?

Is it possible you have +10dB Input Level gain enabled on the RED ONE?

There was a bug in earlier Builds (17 for example) where that was the default setting when the camera turned on....

[Stuart English]

Just received my upgrade and ran a line level test. -20 tone from Sound Devices, line input to Red at 0dB. (the other option is +10)

tone hit the hash mark, maybe a touch hot.....

The test tone reads out as -16 in FCP, which is -28 below unity.

So given that, if the scale logic of the meters in the camera and FCP are the same, you would expect to see a recording of a line level tone input that aligns with the witness mark of a Rev B audio card on the RED ONE (i.e a signal at -20dBFS) lining up to ( 12 - 20 = -8 ) - 8 on the FCP meters.

But a big caveat - this assumes they use the same logic to represent the data.

Does FCP have a 4 offset and 2:1 scale difference ?

[James Brundige]

That suggests your input level is too high or you have +10dB gain enabled for line input.

It shouldn't be lighting up yellow, as you are correct when it changes from green to yellow its indicating about 6dB of remaining headroom before clip...

When I run my Sound Devices 302 (I don't have it with me unfortunately) at -20dBFS, my camera indicates its hitting the witness mark right on the nail.....


What build of firmware are you on?

Is it possible you have +10dB Input Level gain enabled on the RED ONE?

There was a bug in earlier Builds (17 for example) where that was the default setting when the camera turned on....

latest build 20. +10 is NOT enabled.

Running dialogue through a pro mixer at the optimal level for the mixer is indeed too hot for the Red. Could be the difference between +4 dB and 0 dB.

I just shot a performance with a board feed to the Red. Kept it out of the yellow, nicely in the middle of the Red meter. I will be very interested to see how it moves into FCP. After that, I will take time to do some thorough tests. From other posts, I gather the mic level input, which has much more adjustment, may work better. The issue as I see it is lining up the mixer's headroom with the camera. My quick tests with tone have not accomplished that.

[Stuart English]

About FCP, I'm confused about their meter. It tops out at +12, which I think is 0 to the rest of the world. What's dBFS?

Can you post a JPEG of the meter that you are looking at in FCP?

Just so we know that we are talking about the same thing.. I'm looking at this - except its vertical. Range is - 96 to 0